Guitar effects pedals are designed to provide electronic alteration to a guitar signal and are often controlled by a footswitch that routes the signal either through the effect or directly to the output through a clean channel. The goal of this paper is to create a trainable, low-power switch that classifies the incoming guitar signal into two or more playing style classes and routes the input to either the bypass or effect channel, replacing the manual method of clocking the footswitch manually through different playing styles. A training data set consisting of nearly 5 hours of recorded single notes and power chords is collected. The trained neural network algorithm is then able to distinguish between these two playing styles with 94.9% accuracy in the test set. An electronic implementation is designed with a Raspberry Pi Pico, preamplifiers, multiplexers, and an analog distortion circuit that runs a neural network trained using Edge Impulse software to perform classification and signal routing in real-time.
(See document for exact affiliation information.)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
